Single tooth gear hob possible?

Nicholas Carr · Mar 4, 2018

Hello all,

New to hobbing and gear cutting. I need to make a handful of cast iron 48DP 96 tooth worm wheels closely matching the factory accuracy of 100 years ago. My spindle and dividing head are synchronized for hobbing. Before taking the time to make a hardened hob, I would like to try this with a carbide or HSS single point cutter. I haven't done the final calculations yet but the arbor will need to be around 1/2" diameter with a tiny cutter which is the part that has me concerned. I'm thinking 1/8" round carbide blank with a setscrew in the side and one in the rear to keep the bit from pushing in. Does anyone with experience see any issues with this setup? Cutting time is not an issue.

Thanks!

Mike C. · Mar 4, 2018

Couple of things... If you are single pointing it, it's not a hob, it's a form cutter. You just dial the dividing head around and cut each slot individually. Works fine. I have done it on a Bport type vertical, a horizontal mill and even a shaper. Use a tooth form pattern and take as long as you need to grind a properly shaped tool that leaves no light when held against the pattern.

johansen · Mar 4, 2018

Mike C. said:
Couple of things... If you are single pointing it, it's not a hob, it's a form cutter. You just dial the dividing head around and cut each slot individually. Works fine. I have done it on a Bport type vertical, a horizontal mill and even a shaper. Use a tooth form pattern and take as long as you need to grind a properly shaped tool that leaves no light when held against the pattern.

the other way to do it is to fit the single point cutter onto something that is threaded at the same DP as the worm gear. after cutting all the teeth once, rotate the cutter so that it is shifted say 30 degrees, cut the teeth again. then rotate it -30 degrees and do it again, then +/-60 degrees, 90, 120,150, 180. doing this turns your single point cutter into a hob with 12 teeth.. which is probably enough.

since he already has the spindle and dividing head synchronized.. this will only be as difficult as cutting a thread that matches the dp, making a nut that you can bolt the single cutting bit on, and some method to secure the bolt to the thread while also allowing you to adjust it.

if you don't want to cut a thread (15.27 tpi, which is almost exactly .6mm pitch) to match the dp then you will need to rotate the cutter, assuming it is held in the spindle of the machine, move the spindle the correct amount.. 12 times.

Nicholas Carr · Mar 4, 2018

Mike C. said:
Couple of things... If you are single pointing it, it's not a hob, it's a form cutter. You just dial the dividing head around and cut each slot individually. Works fine. I have done it on a Bport type vertical, a horizontal mill and even a shaper. Use a tooth form pattern and take as long as you need to grind a properly shaped tool that leaves no light when held against the pattern.

Hi Mike, cutting individually is another option but I invested time modifying the machine for hobbing when practical so I would like to go that route. I just finished the machine and now thinking about setup. I'm going to cut Delrin first to make sure it all works.

Nicholas Carr · Mar 4, 2018

johansen said:
the other way to do it is to fit the single point cutter onto something that is threaded at the same DP as the worm gear. after cutting all the teeth once, rotate the cutter so that it is shifted say 30 degrees, cut the teeth again. then rotate it -30 degrees and do it again, then +/-60 degrees, 90, 120,150, 180. doing this turns your single point cutter into a hob with 12 teeth.. which is probably enough.

since he already has the spindle and dividing head synchronized.. this will only be as difficult as cutting a thread that matches the dp, making a nut that you can bolt the single cutting bit on, and some method to secure the bolt to the thread while also allowing you to adjust it.

if you don't want to cut a thread (15.27 tpi, which is almost exactly .6mm pitch) to match the dp then you will need to rotate the cutter, assuming it is held in the spindle of the machine, move the spindle the correct amount.. 12 times.

What you wrote sounds like a major point that I missed. I thought a single (correct angle) straight sided cutter would duplicate a hob but just take a little longer to cut. If that's the case I would rather make a proper hob.

HuFlungDung · Mar 4, 2018

If you have synchronized the single point tool with the work then it should be no big deal to take multiple passes (once you have it planned out). You may even be able to shift the tool/work position by indexing either the driver or the driven gear in your gear train, accompanied by a lateral shift in the work position so as to simulate a slight rotation of the workpiece relative to the tool.

I've cut fine pitch worm wheels on a cnc mill using a threading cycle in this manner, merely shifting the start point of the threading cycle in conjunction with a slight adjustment in the rotational position of the indexer. In my case, the descending tool took a chip on as many as three tooth spaces at a time, so I only had to shift the tool and work maybe 4 times in all, before I found finer incremental shifts were not cutting anything detectable.

Nicholas Carr · Mar 4, 2018

HuFlungDung said:
If you have synchronized the single point tool with the work then it should be no big deal to take multiple passes (once you have it planned out). You may even be able to shift the tool/work position by indexing either the driver or the driven gear in your gear train, accompanied by a lateral shift in the work position so as to simulate a slight rotation of the workpiece relative to the tool.

I've cut fine pitch worm wheels on a cnc mill using a threading cycle in this manner, merely shifting the start point of the threading cycle in conjunction with a slight adjustment in the rotational position of the indexer. In my case, the descending tool took a chip on as many as three tooth spaces at a time, so I only had to shift the tool and work maybe 4 times in all, before I found finer incremental shifts were not cutting anything detectable.

It's electronically synchronized with no fine adjustment other than number of teeth, bare bones. I just wish I could visualize in my mind what all of these other cutting surfaces in a hob are doing other than cutting a little from more than one space at a time which I can understand.

I was thinking of plunging in whatever # of thousands, letting the gear make at least one rotation, rinse and repeat until depth of cut is reached.

Peter. · Mar 4, 2018

I do believe I have a 48dp hob that is almost spent, 10mm bore. If I do then you can have it for postage.

Nicholas Carr · Mar 4, 2018

Peter. said:
I do believe I have a 48dp hob that is almost spent, 10mm bore. If I do then you can have it for postage.

Thank you very much for the offer Peter. It would need to be the same diameter as the mating worm for this worm wheel which I haven't calculated yet and mine is missing so I need to make a worm as well. I'm guessing around .500 diameter from pictures I have seen.

Peter. · Mar 4, 2018

Ah, no I missed that requirement. Mine is 1.25" diameter and actually it's brand new - it was the 38dp I have that I was thinking about. Sorry to get your hopes up.

Nicholas Carr · Mar 4, 2018

Peter. said:
Ah, no I missed that requirement. Mine is 1.25" diameter and actually it's brand new - it was the 38dp I have that I was thinking about. Sorry to get your hopes up.

No problem at all. It was nice of you to offer it.

Nicholas Carr · Mar 4, 2018

johansen said:
the other way to do it is to fit the single point cutter onto something that is threaded at the same DP as the worm gear. after cutting all the teeth once, rotate the cutter so that it is shifted say 30 degrees, cut the teeth again. then rotate it -30 degrees and do it again, then +/-60 degrees, 90, 120,150, 180. doing this turns your single point cutter into a hob with 12 teeth.. which is probably enough.

since he already has the spindle and dividing head synchronized.. this will only be as difficult as cutting a thread that matches the dp, making a nut that you can bolt the single cutting bit on, and some method to secure the bolt to the thread while also allowing you to adjust it.

if you don't want to cut a thread (15.27 tpi, which is almost exactly .6mm pitch) to match the dp then you will need to rotate the cutter, assuming it is held in the spindle of the machine, move the spindle the correct amount.. 12 times.

I'm starting to understand this now. What I was going to do would create a tooth form the same shape as my cutter which might be a killer because my poor little bit would be cutting on all three sides at once. But say that it survives the first cut. Could I simply disconnect the electronic sync and either turn the spindle or dividing head and achieve the same results as your screw idea? I think so if I understand this correctly.

I found this article from an older thread here that helped turn the light bulb on.

Gear cutting with a rack form multi-tooth cutter

EPAIII · Mar 4, 2018

Yes, there is a problem with a single cutter being used with the hobbing process. Hobs are cut with the tooth profile of a rack. They have STRAIGHT sides which is what perfect rack teeth would have. They rely on the multiple teeth that are spaced around the hob to successively make small, straight cuts at different angles to GENERATE the proper, involute profile teeth of the various gears of the same pitch. Most hobs will have at least eight teeth spaced around their axis and many will have ten, twelve, or more. The more teeth, the better the approximation to a true involute. And, inversely, the fewer teeth, the worse that approximation will be. With a single cutter, the approximation will be very poor. You are talking about making worm wheels and the error with them will be less, probably a lot less if a single cutter is used. But just how good they will be can only be answered by making one and seeing how it works. The success may depend on weather you are making a simple, helical form worm wheel which provides only a single point of contact for each tooth or if you want a formed worm which will give a line of contact at each tooth. The former will be easier to make while the latter will be more inaccurate with a single cutter tool.

You can easily generate the proper, straight tooth profile on a hob with a lathe tool that is ground much like an Acme threading tool. It would have straight sides, exactly like an Acme threading tool. And the hob is cut like a screw thread on a lathe but you will need individual change gears and a banjo to get the correct pitch. Very few quick change gear boxes will give you the correct pitch setting. But the real problem, in the home or small shop, is generating the relief that is needed on each tooth for proper cutting action. One solution to this has been to use a cam operated tool post that extends and retracts the tool while cutting each tooth.

Or to cut a simple, helical form worm you can just purchase a ready made hob. You talk about a quality level of 100 years ago, whatever that may have been. You can purchase hobs that will probably be OK for your purposes for a fairly low price. If you want a proper, formed worm wheel with line type contact then you will probably have to make your own hob cutter because it must be the same diameter as the worm you will use with that worm wheel.

johansen · Mar 5, 2018

EPAIII said:
If you want a proper, formed worm wheel with line type contact then you will probably have to make your own hob cutter because it must be the same diameter as the worm you will use with that worm wheel.

I think my solution to mount the single point cutter on a threaded rod of 15.27 tpi or .6mm pitch is the easiest solution to solve both problems. the diameter is easily changed (set and forget at the correct value assuming you don't need to resharpen it) and the cutter simply rotated and as its rotated it is shifted axially the correct amount.

however: since the OP has an electronically synchronized system already all this can be done electronically, if the machine is up to the task. for one continuous operation the single point cutter rotating at say 95 rpm, the gear blank rotating at 1 rpm, while the cutter is fed axially at a rate of something like .065" per minute (this makes up the difference and is how you end up with a 96 tooth gear when the ratio of rpm is 95). this will produce on the order of minimum 10 facets per tooth, and the time to make the gear will be on the order of 9 minutes.. because you have to start the cutter about 4 teeth away from the center line, and if you run the program until the cutter exits the gear that's about 8-9 teeth.

johansen · Mar 5, 2018

Nicholas Carr said:
It's electronically synchronized with no fine adjustment other than number of teeth, bare bones. I just wish I could visualize in my mind what all of these other cutting surfaces in a hob are doing other than cutting a little from more than one space at a time which I can understand.

I was thinking of plunging in whatever # of thousands, letting the gear make at least one rotation, rinse and repeat until depth of cut is reached.

Presuming you don't have the ability to set a 95:1 turns ratio while feeding the cutter through the gear blank at .065" per minute.... i suggest you go with my solution of mounting the single point cutter on a threaded rod of 15.27 tpi (15tpi might be good enough) or .6mm pitch. take one cut at center line to the full depth and then rotate the cutter +/-60, +/-120, and +/-180 degrees and you get 6 facets per tooth which is probably good enough. if not then do it again at +/- 270 degrees and +/-360 degrees.

assuming you can run the cutter at 96 rpm and the gear blank at 1 rpm it only takes 1 minute to make each additional pass after plunging to the full depth in the first pass, every time you adjust the cutter +/-60 degrees you're only going to be shaving a small fraction of metal off the gear so you don't need to move the machine's axis to take the cut in multiple steps.. just stop the spindle wherever and adjust the cutter +/-60 degrees and restart it.

SeaMoss · Mar 5, 2018

Straight-sided single tooth cutter will work fine. Hope the table is synchronized as well, OP needs that.

Look up tangential feed hobbing.

Nicholas Carr · Mar 5, 2018

johansen said:
Presuming you don't have the ability to set a 95:1 turns ratio while feeding the cutter through the gear blank at .065" per minute.... i suggest you go with my solution of mounting the single point cutter on a threaded rod of 15.27 tpi (15tpi might be good enough) or .6mm pitch. take one cut at center line to the full depth and then rotate the cutter +/-60, +/-120, and +/-180 degrees and you get 6 facets per tooth which is probably good enough. if not then do it again at +/- 270 degrees and +/-360 degrees.

assuming you can run the cutter at 96 rpm and the gear blank at 1 rpm it only takes 1 minute to make each additional pass after plunging to the full depth in the first pass, every time you adjust the cutter +/-60 degrees you're only going to be shaving a small fraction of metal off the gear so you don't need to move the machine's axis to take the cut in multiple steps.. just stop the spindle wherever and adjust the cutter +/-60 degrees and restart it.

I want to thank you again for pointing my error out. Last night I made a post with a link to an article that I found in a 2010 thread on this forum but the post hasn't shown up yet. Something about a moderator has to approve it probably because of the link.

Anyway, with the help of that article I completely understand what is going on now which was bugging me because I couldn't follow it in my head. I don't want to post the link again but the name is "Gear cutting with a rack form multi-tooth cutter" by Tony Jeffree and he was showing visual examples of what you have pointed out to me.

I could switch it to to 95 teeth but I'm not sure what the results would be as far as holding the correct positioning. The first attempt will be with Delrin no matter how I end up doing it. My thought last night was to temporarily disable the sync and manually adjust the spindle or dividing head without using any sort of screw. He was adjusting both the blank and cutter in his article which makes sense without a screw arrangement so maybe I could give that a try?

At this point I may want to try grinding an involute cutter or make the proper hob. A large part of this project is the learning aspect so time and money isn't an issue. I also now have concerns of that tiny bit surviving while cutting on all three sides at once as it would during the first cut. Again, I'm REALLY glad I posted this question here because I was way off base with my original plan.

Nicholas Carr · Mar 5, 2018

johansen said:
Presuming you don't have the ability to set a 95:1 turns ratio while feeding the cutter through the gear blank at .065" per minute.... i suggest you go with my solution of mounting the single point cutter on a threaded rod of 15.27 tpi (15tpi might be good enough) or .6mm pitch. take one cut at center line to the full depth and then rotate the cutter +/-60, +/-120, and +/-180 degrees and you get 6 facets per tooth which is probably good enough. if not then do it again at +/- 270 degrees and +/-360 degrees.

assuming you can run the cutter at 96 rpm and the gear blank at 1 rpm it only takes 1 minute to make each additional pass after plunging to the full depth in the first pass, every time you adjust the cutter +/-60 degrees you're only going to be shaving a small fraction of metal off the gear so you don't need to move the machine's axis to take the cut in multiple steps.. just stop the spindle wherever and adjust the cutter +/-60 degrees and restart it.

Darn, I just made another post that needs approval from a moderator. I wonder what's going on?

edit: well this post showed up. I have one post from last night and one from this morning trying to answer you.

Nicholas Carr · Mar 5, 2018

Test test..............................

EPAIII said:
Yes, there is a problem with a single cutter being used with the hobbing process. Hobs are cut with the tooth profile of a rack. They have STRAIGHT sides which is what perfect rack teeth would have. They rely on the multiple teeth that are spaced around the hob to successively make small, straight cuts at different angles to GENERATE the proper, involute profile teeth of the various gears of the same pitch. Most hobs will have at least eight teeth spaced around their axis and many will have ten, twelve, or more. The more teeth, the better the approximation to a true involute. And, inversely, the fewer teeth, the worse that approximation will be. With a single cutter, the approximation will be very poor. You are talking about making worm wheels and the error with them will be less, probably a lot less if a single cutter is used. But just how good they will be can only be answered by making one and seeing how it works. The success may depend on weather you are making a simple, helical form worm wheel which provides only a single point of contact for each tooth or if you want a formed worm which will give a line of contact at each tooth. The former will be easier to make while the latter will be more inaccurate with a single cutter tool.

You can easily generate the proper, straight tooth profile on a hob with a lathe tool that is ground much like an Acme threading tool. It would have straight sides, exactly like an Acme threading tool. And the hob is cut like a screw thread on a lathe but you will need individual change gears and a banjo to get the correct pitch. Very few quick change gear boxes will give you the correct pitch setting. But the real problem, in the home or small shop, is generating the relief that is needed on each tooth for proper cutting action. One solution to this has been to use a cam operated tool post that extends and retracts the tool while cutting each tooth.

Or to cut a simple, helical form worm you can just purchase a ready made hob. You talk about a quality level of 100 years ago, whatever that may have been. You can purchase hobs that will probably be OK for your purposes for a fairly low price. If you want a proper, formed worm wheel with line type contact then you will probably have to make your own hob cutter because it must be the same diameter as the worm you will use with that worm wheel.

My longer posts seem to get hung up for moderator approval or lost forever. Let my try adding shorter and saved snippets with the edit feature.

edit:

I need to duplicate the original wheel and it's an enveloping design but the worm is cut straight and not a double if that is what you were referring to.

I do need to make some worms as well. I converted my manual lathe to CNC Back in the 80's. Nothing spectacular but I can cut any lead with it and was planning on using .065

I assume that's for radial relief only and hobs do not have any relief on the sides of the teeth? I watched some videos of home made machines, very cool! And I saw a video of a green machine in England IIRC that was a real hob reliever. He makes gears for a living.

Another feature of my lathe is CNC spindle control by just installing a timing belt which I keep off when not needed. It's not the easiest controller to program but I think I could duplicate the motions of a reliever if I decide to go with a hob.

johansen · Mar 5, 2018

Nicholas Carr said:
I could switch it to to 95 teeth but I'm not sure what the results would be as far as holding the correct positioning. The first attempt will be with Delrin no matter how I end up doing it.

i might be mistaken on adjusting the gear ratio but i hope you understood what i was saying and so just not to confuse anyone in the future:

as the single point cutter makes 96 rotations and the gear blank makes one complete turn, the idea is that you then move the cutter .065" axially and rotate the gear 3.75 degrees (360/96), then run it again. (doing this several times to generate a more accurate profile), but this can be done continuously.

another way to explain it is: the single point cutter has rotated 960 times, the gear has rotated 10 times, plus 9 teeth, and the cutter has moved 0.65 inches axially, passing into and out of mesh with the worm gear. a single point cutter would generate something like 10 facets on each side of the tooth, in theory.. and this might only take 10 minutes to cut the gear.. you just have three moving parts to synchronize instead of two.

Single tooth gear hob possible?

Plastic

Diamond

Stainless

Plastic

Plastic

Diamond

Plastic

Titanium

Plastic

Titanium

Plastic

Plastic

Diamond

Stainless

Stainless

Banned

Plastic

Plastic

Plastic

Stainless

Similar threads